IntroductionHand hygiene is a key component of infection control in healthcare. WHO recommends that healthcare workers perform six specific poses during each hand hygiene action. SureWash (Glanta Ltd, Dublin, Ireland) is a novel device that uses video-measurement technology and immediate feedback to teach this technique. We assessed the impact of self-directed SureWash use on healthcare worker hand hygiene technique and evaluated the device's diagnostic capacity.MethodsA controlled before-after study: subjects in Group A were exposed to the SureWash for four weeks followed by Group B for 12 weeks. Each subject's hand hygiene technique was assessed by blinded observers at baseline (T0) and following intervention periods (T1 and T2). Primary outcome was performance of a complete hand hygiene action, requiring all six poses during an action lasting ≥20 seconds. The number of poses per hand hygiene action (maximum 6) was assessed in a post-hoc analysis. SureWash's diagnostic capacity compared to human observers was assessed using ROC curve analysis.ResultsThirty-four and 29 healthcare workers were recruited to groups A and B, respectively. No participants performed a complete action at baseline. At T1, one Group A participant and no Group B participants performed a complete action. At baseline, the median number of poses performed per action was 2.0 and 1.0 in Groups A and B, respectively (p = 0.12). At T1, the number of poses per action was greater in Group A (post-intervention) than Group B (control): median 3.8 and 2.0, respectively (p<0.001). In Group A, the number of poses performed twelve weeks post-intervention (median 3.0) remained higher than baseline (p<0.001). The area under the ROC curves for the 6 poses ranged from 0.59 to 0.88.DiscussionWhile no impact on complete actions was demonstrated, SureWash significantly increased the number of poses per hand hygiene action and demonstrated good diagnostic capacity.

Mean-shift tracking is a data-driven technique for track-

No abstract

We develop a gradient-based normalised cross-correlation tracker that is as robust as brute-force template matching while being significantly more computationally efficient. The technique serves as the basis of our track validation algorithm: by tracking an object forwards in time, reinitialising at the end of the sequence and then tracking backwards in time, we can determine whether or not the object has been followed correctly -the forwards and backwards trajectories will be very different for tracking failures. If such a failure occurs, we iteratively attempt to validate shorter portions of the video sequence until validation is achieved. The algorithm provides a means of determining whether or not an object was tracked successfully without the need for ground truth data.